Automatic control method and device for double-slag opening furnace front slag discharge of an electric arc furnace

Abstract

The present application relates to the technical field of slagging of an ore smelting furnace, and in particular to a method and device for automatic control of slagging at a double-slagging opening of an ore smelting furnace; the method comprises: collecting high-aluminum slag phase parameters and working condition parameters in real time during smelting of the ore smelting furnace, and outputting a standardized data set after pre-processing; inputting the standardized data set into a prediction model, outputting a prediction result and generating a corresponding grade of a slagging plan; performing differential time-sharing slagging control of the double-slagging opening according to the slagging plan and the slag phase spatial distribution characteristics in the furnace, and synchronously collecting real-time slag phase parameters during the slagging process; through the whole-process automatic control logic of multi-dimensional slag phase and working condition parameter standardized collection, early prediction of high-aluminum slag fluidity, differential time-sharing slagging of the double-slagging opening, and closed-loop linkage of slagging and slag-making systems, the technical problems of uncontrollable slag fluidity, single function of the double-slagging opening, and low slagging efficiency caused by accumulation in the furnace hearth and disconnection between slagging and slag-making are solved.

Description

Technical Field

[0001] This invention relates to the field of slag discharge technology for submerged arc furnaces, and in particular to an automatic control method and device for slag discharge at the furnace front of a double slag outlet in a submerged arc furnace. Background Technology

[0002] In the process of smelting low-nickel ferroalloys and nickel-chromium alloys in submerged arc furnaces of laterite nickel ore, slag removal at the furnace front is a core link to ensure the stable operation of the smelting process, directly affecting the production efficiency, resource recovery rate, and operational safety of the submerged arc furnace. With changes in the global nickel ore supply pattern, the supply of high-grade nickel ore has continued to shrink, and low-grade, high-alumina laterite nickel ore, which is easy to mine and has low procurement costs, has become the mainstream raw material for smelting low-nickel ferroalloys and nickel-chromium alloys in domestic submerged arc furnaces. Currently, the industry generally adopts a double slag outlet structure design for slag removal operations at the furnace front, combined with the principle of three-timed and double-controlled slag removal, and equipped with online monitoring devices for basic operating parameters in the furnace and slag-making system optimization methods to adapt to the basic production needs of high-alumina slag smelting.

[0003] The existing dual-slag-mouth slag discharge technology for electric arc furnaces cannot adapt to the characteristics of raw materials such as large fluctuations and sudden changes in slag viscosity during the smelting of high-alumina slag. Slag discharge control can only rely on manual experience and can not achieve advance prediction and active control of slag flowability. The structural advantages of the dual-slag-mouth have not been effectively explored and utilized. At the same time, the slag discharge operation is completely disconnected from the slag-making system, which can easily lead to problems such as hearth accumulation, poor slag discharge, frequent fluctuations in furnace conditions, and excessive metal loss in slag. As a result, the large-scale smelting of high-alumina, low-grade laterite nickel ore has poor operational stability, high energy consumption, and low resource utilization, making it difficult to meet the needs of the stainless steel industry chain for low-cost and self-controllable raw material smelting. Summary of the Invention

[0004] The purpose of this invention is to provide an automatic control method and device for slag discharge at the front of a double slag outlet in an electric arc furnace, which aims to solve the technical problems of poor operational stability, high energy consumption, and low resource utilization in the large-scale smelting of high-alumina, low-grade laterite nickel ore in the prior art, making it difficult to meet the low-cost and self-controllable raw material smelting needs of the stainless steel industry chain.

[0005] To achieve the above objectives, the present invention provides an automatic control method for slag discharge at the front of a double slag outlet in a submerged arc furnace, comprising the following steps:

[0006] Real-time acquisition of high-alumina slag phase parameters and operating parameters during the smelting process of an electric arc furnace; output of standardized dataset after preprocessing.

[0007] The standardized dataset is input into a pre-trained high-alumina slag fluidity coupled prediction model, which outputs prediction results and generates corresponding slag discharge plans.

[0008] Based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, differentiated time-sharing slag discharge control with dual slag outlets is implemented, and real-time slag phase parameters are collected simultaneously during the slag discharge process.

[0009] The real-time slag phase parameters are compared with the preset target range, and the slag-making system is dynamically adjusted synchronously. The adjusted slag-making parameters are fed back to the initial acquisition stage for the next round of data updates and predictions.

[0010] Among them, in the step of real-time acquisition of high-alumina slag phase parameters and operating condition parameters during the smelting process of the electric arc furnace, and outputting a standardized dataset after preprocessing:

[0011] The data collected included production data from the process of smelting low-nickel iron and nickel-chromium alloys in an electric arc furnace. High-alumina slag phase parameters included slag Al2O3 content, binary basicity, multi-point slag temperature, and multi-point viscosity. Operating parameters included real-time air supply parameters of the electric arc furnace, slag formation parameters, hearth pressure, melt level, and Ni and Cr composition data of the molten iron.

[0012] All collected data are subjected to moving average filtering and time-series alignment processing to ensure that the timestamp error of all data is ≤1s, and a standardized dataset is generated. The standardized dataset is directly used as the input source of the high-alumina slag fluidity coupling prediction model.

[0013] Among them, in the step of real-time acquisition of high-alumina slag phase parameters and operating condition parameters during the smelting process of the electric arc furnace, and outputting a standardized dataset after preprocessing:

[0014] A multi-point temperature measurement array with no less than 8 collection points arranged along the radial and vertical directions of the furnace is used to collect multi-point slag temperature data covering the center and edge areas of the furnace. The multi-point slag temperature data is used to determine the spatial distribution characteristics of the slag phase in the furnace and to monitor the final slag temperature in real time. The final slag temperature control threshold is ≤1600℃.

[0015] An online laser component detection sensor with a detection frequency of no less than once per minute and an Al2O3 content detection error of ≤±0.5% is used to collect slag component data with an Al2O3 content range of 15%-25%. The component data is used for both model input and dynamic adjustment of the slag-forming process.

[0016] In the step of inputting the standardized dataset into a pre-trained high-alumina slag flowability coupled prediction model, outputting prediction results, and generating corresponding slag discharge plans:

[0017] The high-alumina slag fluidity coupling prediction model is a machine learning model pre-trained based on the gradient boosting decision tree algorithm. The input parameters are standardized datasets, and the output parameters are the predicted values ​​of slag viscosity and fluidity change trends for the next 15-30 minutes. At the same time, it combines multi-point slag temperature data to output the spatial distribution characteristics of slag phases in the furnace.

[0018] A first and a second viscosity threshold are preset. The first threshold is less than the second threshold. The viscosity prediction value is compared with the two thresholds in turn, and three levels of prediction results are output: high fluidity, medium fluidity, and low fluidity. Each level of prediction result is matched with the three-time dual-control slag discharge principle, corresponding to a unique slag discharge plan.

[0019] In this process, given a preset viscosity threshold and a second threshold, where the first threshold is less than the second threshold, the predicted viscosity value is compared sequentially with the two thresholds to output three levels of prediction results: high fluidity, medium fluidity, and low fluidity. Each level of prediction result is matched with the three-timed dual-control slag discharge principle, corresponding to a unique slag discharge plan.

[0020] When the viscosity prediction value is lower than the first threshold, the high fluidity prediction result is output, which triggers the regular slag discharge plan, and the interval between the two slag discharges is fixed to be no more than 2 hours.

[0021] When the viscosity prediction value is between the first threshold and the second threshold, the fluidity prediction result is output, which triggers the pre-adjusted slag discharge plan and extends the slag discharge grace period by 30 minutes according to the amount of slag carried by the molten iron.

[0022] When the viscosity prediction value is higher than the second threshold, the low fluidity prediction result is output, which triggers the emergency slag discharge plan. The slag valve opening size is adjusted to match the peak and valley values ​​of the molten iron flow.

[0023] The three types of slag discharge plans serve as the sole basis for subsequent differentiated time-sharing slag discharge control at the dual slag outlets.

[0024] Among them, in the step of implementing differentiated time-sharing slag discharge control at dual slag outlets based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and simultaneously collecting real-time slag phase parameters during the slag discharge process:

[0025] The dual slag inlets of the electric arc furnace are arranged radially and differentiatedly along the furnace chamber. The main slag inlet corresponds to the low-viscosity slag accumulation area in the center of the furnace chamber and is used for routine slag discharge. The backup slag inlet corresponds to the high-viscosity, high-alumina slag accumulation area at the edge of the furnace chamber and is used for maintenance of the main slag inlet and centralized discharge of high-alumina slag. The arrangement of the two slag inlets corresponds one-to-one with the spatial distribution characteristics of the slag phase in the furnace.

[0026] Based on the level of the triggered slag discharge plan, the slag discharge sequence and slag valve opening of the main slag outlet and the backup slag outlet are matched respectively, and slag of corresponding viscosity is discharged in different areas.

[0027] During the slag discharge process, real-time slag phase parameters are collected synchronously according to the collection items and collection frequency of the initially collected high-alumina slag phase parameters. These real-time slag phase parameters are directly used for the dynamic adjustment of the subsequent slag-making process.

[0028] Among them, in the step of matching the slag discharge sequence and slag valve opening of the main slag outlet and the backup slag outlet according to the triggered slag discharge plan level, and discharging slag of corresponding viscosity in different areas:

[0029] When the routine slag discharge plan is triggered, the slag valve opening of the main slag outlet is maintained at 30%-50% for continuous slag discharge, while the standby slag outlet is in a locked state.

[0030] When the pre-adjusted slag discharge plan is triggered, the main slag outlet maintains a 30%-50% opening for continuous slag discharge, and the standby slag outlet is opened once every 30 minutes, with each opening lasting 3-5 minutes, and the slag valve opening is 20%-30%.

[0031] When the emergency slag discharge plan is triggered, the opening of the main slag outlet valve is reduced to 10%-20%, the standby slag outlet is fully opened, and high-viscosity, high-alumina slag is continuously discharged until the viscosity prediction value falls below the second threshold.

[0032] In the steps of comparing the real-time slag phase parameters with the preset target range, simultaneously performing dynamic adjustments to the slag-forming regime, and feeding back the adjusted slag-forming parameters to the initial data acquisition stage for the next round of data updates and predictions:

[0033] The target ranges for Al2O3 content, binary basicity, and viscosity of the slag are preset. The binary basicity is controlled in stages, with targets of 0.9-1.0 for the initial slag, 1.1-1.2 for the intermediate slag, and 1.05-1.15 for the final slag.

[0034] When the real-time Al2O3 content exceeds the target range, an automatic command to adjust the flux addition amount is sent to the slag-forming control system.

[0035] When the real-time binary alkalinity exceeds the target range for the corresponding stage, an automatic lime addition adjustment command is sent to the slag-forming control system.

[0036] When the real-time viscosity exceeds the target range or the real-time slag temperature exceeds 1600℃, the amount of flux and lime added should be adjusted simultaneously until the real-time slag phase parameters fall back to the target range.

[0037] Prior to the step of inputting the standardized dataset into the pre-trained high-alumina slag fluidity coupling prediction model:

[0038] Historical production data of smelting low-nickel iron and nickel-chromium alloys using high-alumina slag in electric arc furnaces were collected. The data collection items of the historical production data were completely consistent with the initial collection parameters. The slag discharge effect data of the corresponding time period were matched, and the model training dataset was constructed after data cleaning and feature engineering.

[0039] The gradient boosting decision tree algorithm is used to train the model. Training parameters and prediction accuracy verification thresholds are set. The trained model is then deployed to the edge computing controller, which synchronously executes the entire process control operation.

[0040] This invention also provides an automatic control device for slag discharge at the furnace front of a double slag outlet in a submerged arc furnace, comprising a data acquisition module, a predictive analysis module, a slag discharge control module, and a closed-loop linkage module; wherein:

[0041] The data acquisition module is used to collect high-alumina slag phase parameters and operating parameters in real time during the smelting process of the electric arc furnace, and output a standardized dataset after preprocessing.

[0042] The predictive analysis module is used to input the standardized dataset into a pre-trained high-alumina slag fluidity coupled prediction model, output prediction results, and generate corresponding slag discharge plans.

[0043] The slag discharge control module is used to perform differentiated time-sharing slag discharge control at the dual slag outlets according to the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and to simultaneously collect real-time slag phase parameters during the slag discharge process.

[0044] The closed-loop linkage module is used to compare the real-time slag phase parameters with the preset target range, synchronously execute dynamic adjustments to the slag-making system, and feed back the adjusted slag-making parameters to the initial acquisition stage for the next round of data updates and predictions.

[0045] This invention discloses an automatic control method and device for slag discharge at the front of a submerged arc furnace with dual slag openings. The method involves real-time acquisition of high-alumina slag phase parameters and operating parameters during the submerged arc furnace smelting process, followed by preprocessing to output a standardized dataset. This standardized dataset is then input into a pre-trained high-alumina slag flowability coupled prediction model, which outputs prediction results and generates a corresponding level of slag discharge plan. Simultaneously, based on the slag discharge plan and the spatial distribution characteristics of the slag phase within the furnace, differentiated time-sharing slag discharge control at both slag openings is implemented, synchronously acquiring real-time slag phase parameters during the slag discharge process. Finally, the real-time slag phase parameters are compared with a preset target range, and dynamic adjustments to the slag-forming regime are synchronously executed. The adjusted slag-forming parameters are fed back to the initial acquisition stage for the next round of data updates and predictions.

[0046] It achieves fully automated pre-control management of slag discharge in front of the furnace of high-alumina slag-type submerged arc furnace. By predicting the fluidity of high-alumina slag in advance, it transforms the traditional passive emergency response into proactive pre-control, fundamentally avoiding problems such as hearth accumulation and poor slag discharge. Through differentiated time-sharing slag discharge at dual slag outlets, it significantly improves the slag discharge efficiency of high-alumina slag and reduces metal loss in the slag. Through the closed-loop linkage between slag discharge and slag making systems, it stabilizes furnace conditions, reduces smelting energy consumption, and can be efficiently adapted to the large-scale smelting of low-grade high-alumina laterite nickel ore, improving resource utilization and reducing the production cost of stainless steel raw materials. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of the automatic control method for slag discharge at the front of the double slag outlet of the electric arc furnace according to the present invention.

[0049] Figure 2 This is a flowchart of steps S100 of the present invention.

[0050] Figure 3 This is a flowchart of steps S200 of the present invention.

[0051] Figure 4 This is a flowchart of steps S202 of the present invention.

[0052] Figure 5 This is a flowchart of steps S300 of the present invention.

[0053] Figure 6 This is a flowchart of steps S302 of the present invention.

[0054] Figure 7 This is a flowchart of steps S400 of the present invention.

[0055] Figure 8 This is a schematic diagram of the automatic slag discharge control device for the double slag outlet of the electric arc furnace of the present invention.

[0056] Figure 9 This is a schematic diagram of the electronic device of the present invention.

[0057] 501 - Data acquisition module, 502 - Predictive analysis module, 503 - Slag discharge control module, 504 - Closed-loop linkage module. Detailed Implementation

[0058] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0059] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0060] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0061] Please see Figures 1-7 This invention provides an automatic control method for slag discharge at the front of a double slag outlet in a submerged arc furnace, comprising the following steps:

[0062] S100: Real-time acquisition of high-alumina slag phase parameters and operating parameters during the smelting process of electric arc furnace, and output of standardized dataset after preprocessing;

[0063] In this embodiment, high-alumina slag phase parameters and operating condition parameters are collected in real time during the smelting process of the electric arc furnace, and a standardized dataset is output after preprocessing. The specific process is as follows:

[0064] S101: The data collected is the production data of the process of smelting low-nickel iron and nickel-chromium alloys in an electric arc furnace. The high-alumina slag phase parameters include slag Al2O3 content, binary basicity, multi-point slag temperature, and multi-point viscosity. The operating parameters include real-time air supply parameters of the electric arc furnace, slag formation parameters, hearth pressure, melt level, and Ni and Cr composition data of molten iron.

[0065] S102: Perform moving average filtering and time-series alignment processing on all collected data to ensure that the timestamp error of all data is ≤1s, and generate a standardized dataset. The standardized dataset is directly used as the input source of the high-alumina slag fluidity coupling prediction model.

[0066] S103: By using a multi-point temperature measurement array with no less than 8 collection points arranged along the radial and vertical directions of the furnace, multi-point slag temperature data covering the center and edge areas of the furnace are collected. The multi-point slag temperature data is used simultaneously for determining the spatial distribution characteristics of the slag phase in the furnace and for real-time monitoring of the final slag temperature. The final slag temperature control threshold is ≤1600℃.

[0067] S104: Using an online laser component detection sensor with a detection frequency of not less than 1 time / minute and an Al2O3 content detection error of ≤±0.5%, slag component data with an Al2O3 content range of 15%-25% are collected. The component data is used for both model input and dynamic adjustment of the slag-forming process.

[0068] Specifically, for the S101 parameters, the collected data covers all dimensions of production data, including slag characteristics, furnace conditions, and smelting objectives, to meet the smelting requirements of high-alumina slag. A specific example: In a 33000kVA submerged arc furnace low-nickel iron smelting production line, the high-alumina slag phase parameters collected in this step are: Al2O3 content 15%-25%, binary basicity 0.9-1.2, slag temperature at multiple points in an 8-point array, and slag viscosity at multiple points. Operating parameters include: real-time airflow 1200-1500 m³ / min, air temperature ≥1200℃, real-time addition of lime / fluorite for slag formation, hearth pressure 0-200 Pa, melt level 800-1200 mm, and molten iron Ni ≥1.5% and Cr ≥15%, fully covering all the data requirements for high-alumina slag smelting.

[0069] For S102 moving average filtering and time series alignment: filter out electromagnetic interference noise on site, unify the time series of multi-source data, and generate a standardized dataset; specific implementation: in the above production line, perform 5-point moving average filtering on the slag temperature data collected by the 8-point temperature measuring array to filter out noise spikes within ±50℃ caused by on-site electrode arcs; for slag temperature data collected once / 10s, composition data once / minute, and operating parameters once / 5s, perform linear interpolation and time series alignment with a time step of 1 minute to ensure that the timestamp error of all data is ≤1s, generate a standardized dataset, and directly connect to the input interface of the subsequent prediction model.

[0070] For the S103 multi-point temperature measurement array: By measuring the temperature at multiple points throughout the furnace, the spatial distribution of the slag phase and the final slag temperature are simultaneously determined. Specifically, in the furnace of the aforementioned production line, four sampling points are arranged radially (corresponding to the furnace center, 1 / 2 radius, 3 / 4 radius, and furnace edge, respectively), and two sampling points are arranged vertically (corresponding to the upper and lower parts of the slag layer, respectively), totaling eight sampling points in the temperature measurement array, with a sampling frequency of once per 10 seconds. Based on the temperature gradient of these eight points, the furnace center temperature of 1550-1600℃ is determined to be a low-viscosity slag accumulation zone, and the furnace edge temperature of 1450-1500℃ is determined to be a high-viscosity, high-alumina slag accumulation zone. Simultaneously, through the temperature measurement data of the entire area, the final slag temperature is stably controlled within 1580℃, meeting the preset threshold requirements.

[0071] For the S104 online laser component detection sensor: it detects slag components in real time, providing data support for model prediction and slag-forming adjustment. Specifically, an online laser component detection sensor is installed at the slag outlet of the furnace in the aforementioned production line. The detection frequency is set to 1 time / minute, with an Al2O3 content detection error ≤ ±0.4%. It can collect slag component data with an Al2O3 content of 15%-25% in real time. The collected component data is directly input into the high-alumina slag fluidity coupling prediction model as the core input feature of the model, and also directly connected to the slag-forming control system as the core basis for adjusting the flux and lime addition amounts.

[0072] S200: Input the standardized dataset into the pre-trained high-alumina slag flowability coupled prediction model, output the prediction results and generate the corresponding level of slag discharge plan;

[0073] In this embodiment, the standardized dataset is input into a pre-trained high-alumina slag flowability coupled prediction model, which outputs prediction results and generates corresponding slag discharge plans. The specific process is as follows:

[0074] S201: The high-alumina slag fluidity coupling prediction model is a machine learning model pre-trained based on the gradient boosting decision tree algorithm. The input parameters are standardized datasets, and the output parameters are the predicted values ​​of slag viscosity and fluidity change trends for the next 15-30 minutes. At the same time, it combines multi-point slag temperature data to output the spatial distribution characteristics of slag phases in the furnace.

[0075] S202: Preset a first threshold and a second threshold for viscosity. The first threshold is less than the second threshold. Compare the predicted viscosity value with the two thresholds in turn and output three levels of prediction results: high fluidity, medium fluidity, and low fluidity. Each level of prediction result is matched with the three-time dual-control slag discharge principle and corresponds to a unique slag discharge plan.

[0076] S2021: When the viscosity prediction value is lower than the first threshold, output the high fluidity prediction result, which triggers the regular slag discharge plan, and the interval between the two slag discharges is fixed to be no more than 2 hours.

[0077] S2022: When the viscosity prediction value is between the first threshold and the second threshold, the flowability prediction result is output, which triggers the pre-adjustment slag discharge plan and extends the slag discharge grace period by 30 minutes according to the amount of slag carried by the molten iron.

[0078] S2023: When the viscosity prediction value is higher than the second threshold, output the low fluidity prediction result, which triggers the emergency slag discharge plan and matches the peak and valley values ​​of molten iron flow by adjusting the slag valve opening size.

[0079] S2024: The three types of slag discharge plans serve as the sole basis for subsequent differentiated time-sharing slag discharge control at the dual slag outlets.

[0080] Specifically, for the S201 gradient boosting decision tree algorithm model: the GBDT algorithm adapted to industrial scenarios is adopted to fit the coupling relationship between multiple parameters and slag viscosity, and simultaneously output the flowability prediction results and slag phase spatial distribution characteristics. In a specific implementation: in the aforementioned 33000kVA submerged arc furnace production line, the pre-trained GBDT model's input parameters are the standardized dataset generated by S100, and the output parameters are the predicted slag viscosity value and flowability trend for the next 20 minutes. Simultaneously, it combines 8 points of slag temperature data to output the spatial distribution characteristics of low viscosity at the center and high viscosity at the edges of the slag phase within the furnace. The model is set with a learning rate of 0.05, a maximum number of iterations of 200, and a tree depth of 6. In field testing, the mean square error between the predicted and actual viscosity values ​​is ≤0.05 Pa·s, and the prediction accuracy is 92.3%, fully meeting the field pre-control requirements.

[0081] For the S202-S2024 dual-threshold three-level classification prediction matching "three-timed dual-control" principle: set dual thresholds to achieve three-level flowability classification, and combine mature industry slag discharge principles to generate a one-to-one corresponding graded slag discharge plan; specific implementation: in the above production line, the preset first viscosity threshold is 0.3 Pa·s, and the second threshold is 0.8 Pa·s; when the model predicts that the viscosity will be 0.2 Pa·s in the next 20 minutes (below the first threshold), the high flowability prediction result is output, triggering the conventional slag discharge plan, strictly following the "three-timed" principle, fixing the interval between two slag discharges at 1.5 hours, not exceeding 2 hours; when When the predicted viscosity is 0.5 Pa·s (between two thresholds), the output is a medium fluidity prediction result, triggering a pre-adjusted slag discharge plan. Following the "dual control" principle, the slag discharge grace period is extended by 30 minutes based on the amount of slag carried by the molten iron, while the slag valve opening is adjusted in advance. When the predicted viscosity is 1.1 Pa·s (above the second threshold), the output is a low fluidity prediction result, triggering an emergency slag discharge plan. By adjusting the slag valve opening size to match the peak and valley values ​​of the molten iron flow, priority is given to ensuring the smooth discharge of high-viscosity slag. All three plans are directly issued to the dual-slag-port actuators, serving as the sole basis for slag discharge control, and can be automatically executed without manual intervention.

[0082] S300: Based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, implement differentiated time-sharing slag discharge control at the dual slag outlets, and simultaneously collect real-time slag phase parameters during the slag discharge process;

[0083] In this embodiment, based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, differentiated time-sharing slag discharge control with dual slag outlets is implemented, and real-time slag phase parameters are collected simultaneously during the slag discharge process. The specific process is as follows:

[0084] S301: The dual slag inlets of the electric arc furnace are arranged radially and differentiatedly along the furnace chamber. The main slag inlet corresponds to the low-viscosity slag accumulation area in the center of the furnace chamber and is used for routine slag discharge. The backup slag inlet corresponds to the high-viscosity, high-alumina slag accumulation area at the edge of the furnace chamber and is used for maintenance of the main slag inlet and centralized discharge of high-alumina slag. The arrangement of the two slag inlets corresponds one-to-one with the spatial distribution characteristics of the slag phase in the furnace.

[0085] S302: Based on the triggered slag discharge plan level, match the slag discharge sequence and slag valve opening of the main slag outlet and the backup slag outlet respectively, and discharge slag of corresponding viscosity in different areas.

[0086] S3021: When the regular slag discharge plan is triggered, the opening of the slag valve at the main slag outlet is maintained at 30%-50% for continuous slag discharge, and the standby slag outlet is in a locked state.

[0087] S3022: When the pre-adjusted slag discharge plan is triggered, the main slag outlet maintains a 30%-50% opening for continuous slag discharge, and the standby slag outlet is opened once every 30 minutes, with each opening lasting 3-5 minutes, and the slag valve opening is 20%-30%.

[0088] S3023: When the emergency slag discharge plan is triggered, the opening of the main slag outlet valve is reduced to 10%-20%, the standby slag outlet is fully opened, and high-viscosity, high-alumina slag is continuously discharged until the viscosity prediction value falls below the second threshold.

[0089] S303: During the slag discharge process, real-time slag phase parameters are collected synchronously according to the collection items and collection frequency of the initially collected high-alumina slag phase parameters. The real-time slag phase parameters are directly used for the dynamic adjustment of the subsequent slag-making process.

[0090] Specifically, regarding the differentiated arrangement of the S301 dual slag outlets: based on the spatial distribution characteristics of high-alumina slag in the furnace, the positions of the main and backup slag outlets are differentiated to accommodate both routine slag discharge, backup in case of failure, and centralized discharge of high-alumina slag. In a specific embodiment: in the aforementioned 33000kVA submerged arc furnace production line, the dual slag outlets are arranged radially along the furnace. The main slag outlet is located at the radial center of the furnace, corresponding to the low-viscosity slag accumulation area at 1550-1600℃ in the furnace center, and is used for routine slag discharge. The backup slag outlet is located at 3 / 4 of the radial radius of the furnace, corresponding to the high-viscosity, high-alumina slag accumulation area at 1450-1500℃ at the furnace edge. This backup slag outlet can be activated during maintenance or failure of the main slag outlet to ensure continuous production, and can also be used for centralized discharge of high-alumina slag at the edge. The arrangement of the two slag outlets completely corresponds to the slag phase spatial distribution characteristics output by the S200 model.

[0091] For the differentiated control matching of the S302-S3023 graded pre-discharge plans: the slag discharge plan level is bound to the control parameters of the dual slag outlets one by one to achieve precise adaptation between slag discharge control and slag flowability; specific implementation: in the above production line, when the conventional slag discharge plan is triggered, the slag valve opening of the main slag outlet is stabilized at 40% for continuous slag discharge, and the standby slag outlet is completely closed, discharging only the low-viscosity slag in the center to ensure stable and continuous daily production; when the pre-adjusted slag discharge plan is triggered, the main slag outlet maintains a 40% opening for continuous slag discharge, and the standby slag outlet is opened once every 30 minutes, with a single opening duration of 4 minutes and a slag valve opening of 25%, discharging high-alumina slag from the edge area in stages to avoid high-alumina slag accumulation and reduce metal entrainment in the slag; when the emergency slag discharge plan is triggered, the slag valve opening of the main slag outlet is reduced to 15% to reduce the discharge of low-viscosity slag, and the standby slag outlet is fully opened to continuously discharge the high-viscosity, high-alumina slag accumulated at the edge until the viscosity predicted by the model drops below 0.8 Pa·s, completely solving the problem of high-alumina slag accumulation.

[0092] For S303 synchronous acquisition of real-time slag phase parameters: real-time data of the slag discharge process is acquired synchronously according to the initial acquisition standard to ensure the consistency and closed-loop nature of the data throughout the process; specific implementation: during the slag discharge process of the above production line, real-time slag phase parameters are acquired synchronously according to the acquisition items and acquisition frequency of S100, including 8-point array slag temperature (1 time / 10s), Al2O3 content and binary basicity (1 time / minute), real-time slag viscosity, slag forming parameters, etc. The acquired real-time data is directly connected to the subsequent slag forming control system as a real-time basis for adjusting the amount of lime and fluxing agent added, and is simultaneously fed back to the data acquisition stage and incorporated into the next round of standardized dataset for continuous iterative optimization of the model.

[0093] S400: Compare the real-time slag phase parameters with the preset target range, and simultaneously perform dynamic adjustment of the slag-making system. The adjusted slag-making parameters are fed back to the initial acquisition stage for the next round of data update and prediction.

[0094] In this embodiment, the real-time slag phase parameters are compared with a preset target range, and the slag-forming regime is dynamically adjusted synchronously. The adjusted slag-forming parameters are fed back to the initial data acquisition stage for the next round of data updates and predictions. The specific process is as follows:

[0095] S401: Preset target ranges for Al2O3 content, binary basicity, and viscosity in slag. The binary basicity is controlled in stages, with targets of 0.9-1.0 for initial slag, 1.1-1.2 for intermediate slag, and 1.05-1.15 for final slag.

[0096] S402: When the real-time Al2O3 content exceeds the target range, an adjustment command for the amount of flux added is automatically sent to the slag-forming control system.

[0097] S403: When the real-time binary alkalinity exceeds the target range of the corresponding stage, it automatically sends a lime addition adjustment command to the slag-forming control system.

[0098] S404: When the real-time viscosity exceeds the target range or the real-time slag temperature exceeds 1600℃, adjust the amount of flux and lime added simultaneously until the real-time slag phase parameters fall back to the target range.

[0099] Specifically, regarding the phased binary basicity control of S401: based on the process requirements of different stages throughout the smelting cycle, phased target ranges for slag parameters are set to achieve precise control throughout the entire cycle. A specific example: in the aforementioned 33000kVA submerged arc furnace low-nickel iron smelting production line, the preset slag target range is: Al2O3 content 18%-20%, phased binary basicity control: initial smelting stage (0-2 hours) 0.9-1.0, middle smelting stage (2-6 hours) 1.1-1.2, final smelting stage (6-8 hours) 1.05-1.15, slag viscosity 0.2-0.5 Pa·s, and final slag temperature ≤1600℃. This target range fully adapts to the process requirements of the entire low-nickel iron smelting cycle, ensuring that the slag characteristics at different smelting stages are in optimal condition.

[0100] For S402-S404, the slag-forming parameters are dynamically adjusted: based on the deviation between the real-time slag phase parameters and the target range, the amount of flux and lime added is precisely adjusted to achieve closed-loop control of slag characteristics; specific example: during the slag discharge process of the above production line, when the real-time detected Al2O3 content rises to 23%, exceeding the target range of 18%-20%, an instruction is automatically sent to the slag-forming control system to increase the amount of fluorite flux added from 80kg / h to 120kg / h, thereby reducing the slag viscosity; When the real-time monitored binary basicity of the smelting process drops to 1.02, which is below the target range of 1.1-1.2, the lime addition is automatically increased from 300 kg / h to 350 kg / h to increase the slag basicity. When the real-time viscosity rises to 0.9 Pa·s and the slag temperature rises to 1620℃, which exceeds the target range, the addition of fluorite and lime is adjusted simultaneously. Within 12 minutes, the slag viscosity is reduced back to 0.45 Pa·s, and the final slag temperature is stabilized at 1580℃, returning completely to the target range.

[0101] Regarding the feedback of slag-forming parameters to the initial data acquisition stage: The adjusted slag-forming parameters are fed back to the data acquisition terminal and incorporated into the next round of prediction input, forming a closed-loop control logic for the entire process; Specific example: In the above production line, the adjusted slag-forming parameters, such as the fluorite addition of 120 kg / h and the lime addition of 350 kg / h, are simultaneously fed back to the S100 data acquisition stage and incorporated into the next round of standardized dataset. This serves as the input feature for the high-alumina slag fluidity coupling prediction model. The model can predict in advance the effect of slag-forming adjustments on slag viscosity, correct the slag discharge plan in advance, and achieve closed-loop adaptive optimization of the entire process.

[0102] Corresponding to the aforementioned embodiments of the automatic slag discharge control method for double slag outlets in submerged arc furnaces, this application also provides embodiments of an automatic slag discharge control device for double slag outlets in submerged arc furnaces.

[0103] Figure 8 This is a block diagram illustrating an automatic control device for slag discharge at the front of a double slag outlet in a submerged arc furnace, according to an exemplary embodiment. (Refer to...) Figure 8 The system may include a data acquisition module 501, a predictive analysis module 502, a slag discharge control module 503, and a closed-loop linkage module 504, wherein:

[0104] The data acquisition module 501 is used to collect high-alumina slag phase parameters and operating condition parameters in real time during the smelting process of the electric arc furnace, and output a standardized dataset after preprocessing.

[0105] The prediction and analysis module 502 is used to input the standardized dataset into a pre-trained high-alumina slag fluidity coupled prediction model, output prediction results, and generate a corresponding level of slag discharge plan.

[0106] The slag discharge control module 503 is used to perform differentiated time-sharing slag discharge control at the dual slag outlets according to the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and to simultaneously collect real-time slag phase parameters during the slag discharge process.

[0107] The closed-loop linkage module 504 is used to compare the real-time slag phase parameters with the preset target range, synchronously execute dynamic adjustments to the slag-making system, and feed back the adjusted slag-making parameters to the initial acquisition stage for the next round of data updates and predictions.

[0108] In this embodiment, the data acquisition module 501 is used to collect high-alumina slag phase parameters and operating condition parameters in real time during the smelting process of the submerged arc furnace, and output a standardized dataset after preprocessing; the prediction and analysis module 502 is used to input the standardized dataset into a pre-trained high-alumina slag flowability coupled prediction model, output prediction results, and generate a corresponding level of slag discharge plan; the slag discharge control module 503 is used to execute differentiated time-sharing slag discharge control at the dual slag outlets according to the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and synchronously collect real-time slag phase parameters during the slag discharge process; the closed-loop linkage module 504 is used to compare the real-time slag phase parameters with a preset target range, synchronously execute dynamic adjustment of the slag-forming system, and feed back the adjusted slag-forming parameters to the initial acquisition stage for the next round of data update and prediction.

[0109] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0110] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0111] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the above-described automatic slag discharge control method for a dual-slag-mouth submerged arc furnace. Figure 9 The diagram shown is a hardware structure diagram of any device with data processing capabilities, including an automatic slag discharge control device for a double-slag-mouth submerged arc furnace provided in an embodiment of the present invention. Except for... Figure 9 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0112] Accordingly, this application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the aforementioned automatic slag discharge control method for a dual-slag-mouth submerged arc furnace. The computer-readable storage medium can be an internal storage unit of any data-processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data-processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data-processing device, and can also be used to temporarily store data that has been output or will be output.

[0113] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0114] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. An automatic control method for slag discharge at the front of a double slag outlet in a submerged arc furnace, characterized in that, Includes the following steps: Real-time acquisition of high-alumina slag phase parameters and operating parameters during the smelting process of an electric arc furnace; output of standardized dataset after preprocessing. The standardized dataset is input into a pre-trained high-alumina slag fluidity coupled prediction model, which outputs prediction results and generates corresponding slag discharge plans. Based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, differentiated time-sharing slag discharge control with dual slag outlets is implemented, and real-time slag phase parameters are collected simultaneously during the slag discharge process. The real-time slag phase parameters are compared with the preset target range, and the slag-making system is dynamically adjusted synchronously. The adjusted slag-making parameters are fed back to the initial acquisition stage for the next round of data updates and predictions.

2. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 1, characterized in that, In the step of real-time acquisition of high-alumina slag phase parameters and operating condition parameters during the smelting process of an electric arc furnace, and outputting a standardized dataset after preprocessing: The data collected included production data from the process of smelting low-nickel iron and nickel-chromium alloys in an electric arc furnace. High-alumina slag phase parameters included slag Al2O3 content, binary basicity, multi-point slag temperature, and multi-point viscosity. Operating parameters included real-time air supply parameters of the electric arc furnace, slag formation parameters, hearth pressure, melt level, and Ni and Cr composition data of the molten iron. All collected data are subjected to moving average filtering and time-series alignment processing to ensure that the timestamp error of all data is ≤1s, and a standardized dataset is generated. The standardized dataset is directly used as the input source of the high-alumina slag fluidity coupling prediction model.

3. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 2, characterized in that, In the step of real-time acquisition of high-alumina slag phase parameters and operating condition parameters during the smelting process of an electric arc furnace, and outputting a standardized dataset after preprocessing: A multi-point temperature measurement array with no less than 8 collection points arranged along the radial and vertical directions of the furnace is used to collect multi-point slag temperature data covering the center and edge areas of the furnace. The multi-point slag temperature data is used to determine the spatial distribution characteristics of the slag phase in the furnace and to monitor the final slag temperature in real time. The final slag temperature control threshold is ≤1600℃. An online laser component detection sensor with a detection frequency of no less than once per minute and an Al2O3 content detection error of ≤±0.5% is used to collect slag component data with an Al2O3 content range of 15%-25%. The component data is used for both model input and dynamic adjustment of the slag-forming process.

4. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 1, characterized in that, In the step of inputting the standardized dataset into the pre-trained high-alumina slag flowability coupled prediction model, outputting the prediction results, and generating the corresponding level of slag discharge plan: The high-alumina slag fluidity coupling prediction model is a machine learning model pre-trained based on the gradient boosting decision tree algorithm. The input parameters are standardized datasets, and the output parameters are the predicted values ​​of slag viscosity and fluidity change trends for the next 15-30 minutes. At the same time, it combines multi-point slag temperature data to output the spatial distribution characteristics of slag phases in the furnace. A first and a second viscosity threshold are preset. The first threshold is less than the second threshold. The viscosity prediction value is compared with the two thresholds in turn, and three levels of prediction results are output: high fluidity, medium fluidity, and low fluidity. Each level of prediction result is matched with the three-time dual-control slag discharge principle, corresponding to a unique slag discharge plan.

5. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 4, characterized in that, Given a preset viscosity first threshold and a second threshold, where the first threshold is less than the second threshold, the predicted viscosity value is compared sequentially with the two thresholds to output three levels of prediction results: high fluidity, medium fluidity, and low fluidity. Each level of prediction result is matched with the three-time dual-control slag discharge principle, corresponding to a unique slag discharge plan in the following steps: When the viscosity prediction value is lower than the first threshold, the high fluidity prediction result is output, which triggers the regular slag discharge plan, and the interval between the two slag discharges is fixed to be no more than 2 hours. When the viscosity prediction value is between the first threshold and the second threshold, the fluidity prediction result is output, which triggers the pre-adjusted slag discharge plan and extends the slag discharge grace period by 30 minutes according to the amount of slag carried by the molten iron. When the viscosity prediction value is higher than the second threshold, the low fluidity prediction result is output, which triggers the emergency slag discharge plan. The slag valve opening size is adjusted to match the peak and valley values ​​of the molten iron flow. The three types of slag discharge plans serve as the sole basis for subsequent differentiated time-sharing slag discharge control at the dual slag outlets.

6. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 1, characterized in that, In the step of implementing differentiated time-sharing slag discharge control at dual slag outlets based on the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and simultaneously collecting real-time slag phase parameters during the slag discharge process: The dual slag inlets of the electric arc furnace are arranged radially and differentiatedly along the furnace chamber. The main slag inlet corresponds to the low-viscosity slag accumulation area in the center of the furnace chamber and is used for routine slag discharge. The backup slag inlet corresponds to the high-viscosity, high-alumina slag accumulation area at the edge of the furnace chamber and is used for maintenance of the main slag inlet and centralized discharge of high-alumina slag. The arrangement of the two slag inlets corresponds one-to-one with the spatial distribution characteristics of the slag phase in the furnace. Based on the level of the triggered slag discharge plan, the slag discharge sequence and slag valve opening of the main slag outlet and the backup slag outlet are matched respectively, and slag of corresponding viscosity is discharged in different areas. During the slag discharge process, real-time slag phase parameters are collected synchronously according to the collection items and collection frequency of the initially collected high-alumina slag phase parameters. These real-time slag phase parameters are directly used for the dynamic adjustment of the subsequent slag-making process.

7. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 6, characterized in that, In the steps of matching the slag discharge sequence and slag valve opening of the main slag outlet and the backup slag outlet according to the triggered slag discharge plan level, and discharging slag of corresponding viscosity in different areas: When the routine slag discharge plan is triggered, the slag valve opening of the main slag outlet is maintained at 30%-50% for continuous slag discharge, while the standby slag outlet is in a locked state. When the pre-adjusted slag discharge plan is triggered, the main slag outlet maintains a 30%-50% opening for continuous slag discharge, and the standby slag outlet is opened once every 30 minutes, with each opening lasting 3-5 minutes, and the slag valve opening is 20%-30%. When the emergency slag discharge plan is triggered, the opening of the main slag outlet valve is reduced to 10%-20%, the standby slag outlet is fully opened, and high-viscosity, high-alumina slag is continuously discharged until the viscosity prediction value falls below the second threshold.

8. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 1, characterized in that, In the steps of comparing the real-time slag phase parameters with the preset target range, synchronously performing dynamic adjustments to the slag-forming regime, and feeding the adjusted slag-forming parameters back to the initial data acquisition stage for the next round of data updates and predictions: The target ranges for Al2O3 content, binary basicity, and viscosity of the slag are preset. The binary basicity is controlled in stages, with targets of 0.9-1.0 for the initial slag, 1.1-1.2 for the intermediate slag, and 1.05-1.15 for the final slag. When the real-time Al2O3 content exceeds the target range, an automatic command to adjust the flux addition amount is sent to the slag-forming control system. When the real-time binary alkalinity exceeds the target range for the corresponding stage, an automatic lime addition adjustment command is sent to the slag-forming control system. When the real-time viscosity exceeds the target range or the real-time slag temperature exceeds 1600℃, the amount of flux and lime added should be adjusted simultaneously until the real-time slag phase parameters fall back to the target range.

9. The automatic control method for slag discharge at the front of the double slag outlet of a submerged arc furnace as described in claim 1, characterized in that, Before inputting the standardized dataset into the pre-trained high-alumina slag flowability coupled prediction model: Historical production data of smelting low-nickel iron and nickel-chromium alloys using high-alumina slag in electric arc furnaces were collected. The data collection items of the historical production data were completely consistent with the initial collection parameters. The slag discharge effect data of the corresponding time period were matched, and the model training dataset was constructed after data cleaning and feature engineering. The gradient boosting decision tree algorithm is used to train the model. Training parameters and prediction accuracy verification thresholds are set. The trained model is then deployed to the edge computing controller, which synchronously executes the entire process control operation.

10. An automatic control device for slag discharge at the front of a double-slag-mouth submersible furnace, employing the automatic control method for slag discharge at the front of a double-slag-mouth submersible furnace as described in claim 1, characterized in that... It includes a data acquisition module, a predictive analysis module, a slag discharge control module, and a closed-loop linkage module; among which: The data acquisition module is used to collect high-alumina slag phase parameters and operating parameters in real time during the smelting process of the electric arc furnace, and output a standardized dataset after preprocessing. The predictive analysis module is used to input the standardized dataset into a pre-trained high-alumina slag fluidity coupled prediction model, output prediction results, and generate corresponding slag discharge plans. The slag discharge control module is used to perform differentiated time-sharing slag discharge control at the dual slag outlets according to the slag discharge plan and the spatial distribution characteristics of the slag phase in the furnace, and to simultaneously collect real-time slag phase parameters during the slag discharge process. The closed-loop linkage module is used to compare the real-time slag phase parameters with the preset target range, synchronously execute dynamic adjustments to the slag-making system, and feed back the adjusted slag-making parameters to the initial acquisition stage for the next round of data updates and predictions.

Images